This application builds on the investigators'prior work assessing and training more effective coping skills in women with an alcoholic partner who is not in treatment. To date, that work has focused largely on the effects of Coping Skills Training (CST) and an alternate, 12-step Facilitation (TSF) treatment on the woman's own functioning and, secondarily, on the partner's drinking. However, exploratory analyses suggest that CST may be particularly effective (relative to TSF) in reducing partner physical violence against the woman, reducing violent-partner drinking over time, and eliminating the positive relationship between partner drinking and violence during follow-up. In this application, we use improved relationship-violence methodology to (a) test the replicability of these findings, (b) evaluate the effects of the treatments on the woman's own violence toward her partner-heretofore not assessed, and (c) explore the constructs and putative causal pathways operating in a heuristic model of alcoholic partner and spouse negative affect, partner alcohol involvement, aversive marital behaviors, and spouse self-control (coping) skills. With respect to the latter, we will explore whether an increase in the woman's coping skill in CST, relative to TSF, will moderate (buffer) the effects of partner alcohol involvement and partner violence on her own negative affect. Skill level also is thought to further moderate the effect of the woman's negative affect on her own violent relationship behaviors. A reduction in the woman's own violence is hypothesized to reduce partner violence and the partner's own negative affect. A reduction in the latter then is thought to account for a reduction in his drinking and a further reduction in his violence against the woman. The above a priori and exploratory aims will be evaluated in a 2- group CST vs. TSF longitudinal design. One hundred and fifty women with a physically-violent alcoholic partner not currently in treatment will be randomly assigned within therapy groups to either CST or TSF. All participants will be followed at 90-day intervals for a period of 12 months posttreatment during which both the woman's and partner's negative affect and aversive marital behaviors are assessed, partner drinking measured, and the woman's coping skill acquisition evaluated. Estimates suggest that nearly half of all women with alcoholic partners experience some partner physical violence. Although these women typically have been advised by family, friends, and others to leave the relationship, many remain. Helping these individuals improve their own functioning and reduce the violence they experience may have a greater public health benefit than simply advising them to leave the partner or referring them to 12-step groups.